 Blackwell Publishers Ltd. 1999, 108 Cowley Road, Oxford, OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA. Mind & Language, ISSN: 0268-1064 Vol. 14. No. 1 March 1999, pp 80–112. Adaptive Modelling and Mindreading DONALD M. PETERSON AND KEVIN J. RIGGS Abstract: This paper sets out to give sufficient detail to the notion of mental simul- ation to allow an appraisal of its contribution to ‘mindreading’ in the context of the ‘false-belief tasks’ used in developmental psychology. We first describe the reasoning strategy of ‘modified derivation’, which supports counterfactual reasoning. We then give an analysis of the logical structure of the standard false-belief tasks. We then show how modified derivation can be used in a hybrid strategy for mindreading in these tasks. We then formulate the hypothesis that children become able (around the age of 4 years) to pass these tasks when they acquire the counterfactual reasoning ability required in this strategy. We find this to be consistent with various empirical findings. We then discuss the implications for the theory/simulation debate, meta- representation and pretence. 1. Introduction How do we gain understanding of other minds? How do we come to know what other people think? How do we predict other people’s opinions and actions? These issues have recently been grouped under the heading of ‘min- dreading’, and are basic to the theory of how we understand, interact with, empathize with and cooperate with one another. To perform mindreading is to gain knowledge of the inaccessible: unless we are telepathic, we cannot simply see into other people’s minds, and technology has not provided us with a ‘cognoscope’ to perform the task. In this paper, we focus on a hybrid strategy for the acquisition of such knowledge through adaptive modelling, and we apply this analysis to a particular task of mindreading used in devel- opmental psychology, Wimmer and Perner’s ‘Maxi Task’. 2. Knowledge of Inaccessible Things It happens in many contexts that we want to answer questions about inac- cessible things: situations in the future, in the past or in inaccessible places, For their helpful comments on previous drafts of this paper, we are grateful to George Botterill, Adrian Hunt and Josef Perner. Some of the theoretical arguments presented here appear in earlier form, in terms of ‘subtractive reasoning’, in Peterson and Riggs (in press), and in briefer form in Riggs and Peterson (in press). Address for correspondence: Donald M. Peterson, Cognitive Science Research Centre, School of Computer Science, University of Birmingham, Birmingham B15 2TT, UK. Email: d.m.petersonKcs.bham.ac.uk.